The IF substitution method of microwave attenuation or insertion loss measurement is very well known. In its broadest sense it involves converting a microwave test signal to some lower or intermediate frequency (IF), such as by heterodyning, so that the converted IF signal frequency remains the same for all measurements over the microwave frequency range of interest. An attenuation measurement at any one microwave frequency requires a first reading of the IF signal level with the device under test shorted out, and a second reading with the device under test in the microwave test signal path. The difference between these readings is a measure of the attenuation of the device under test. Utilization of a constant IF signal frequency avoids the problems inherent in using measuring arcuitry which inherently has frequency-dependent characteristics. However, a major problem with this basic approach is that it is limited to a small range of attenuation measurements by the gain limitations of any practical IF amplifier employed in the measurement circuitry. For example, it is desirable to permit attenuation measurements from 0 to 120 dB which corresponds to a range of one million to one. Early on it was suggested that a step attenuator be employed in series with the IF amplifier to reduce the amplifier gain requirements. However, even in such instances it was found that the amplifier characteristics tend to vary to an unacceptable degree.
In U.S. Pat. Nos. 3,034,045 (Weinschel) and 3,104,354 (Weinschel et al), there is disclosed an IF substitution technique which eliminates the critical aspects of the IF amplifier. In this arrangement, the IF signal derived from the microwave test signal and a reference IF signal of the same frequency, derived from a standard source, are fed in interlaced fashion to an IF amplifier. Any difference in amplitude between the two IF signals results in a square wave output from the IF amplifier at the interlacing frequency, nominally 1 KHz. A precision step attenuator, connected in series with the reference IF signal, is adjusted to equalize the two IF signal amplitudes and eliminate the square. The difference between the precision attenuator readings with the device under test in and out of the circuit is a measure of the attenuation of the device under test. During these readings, the gain characteristic of the IF amplifier is not critical since the only purpose of the amplifier is to detect the existence of a difference between the two IF signal levels. However, even with this sophisticated and widely accepted approach to the IF substitution technique, it has been found that 0 to 90 dB is the practical range of permitted attenuation measurement. The reason for the limitation is that below 90 dB system noise starts to interfere with measurement accuracy. The noise derives in part from the reference attenuator and in part from the switching required to interlace the IF signals. Some of this noise can be eliminated if turning the signal sources on and off rather than switching them through gates to achieve interlacing. However, many such sources do not operate well, both in terms of frequency stability and longevity, when repetitively turned on and off.
Another prior art IF substitution approach is described in a paper by Little et al entitled "An NBS Developed Network Analyzer" which appeared in the proceedings from the Conference on Precision Electromagnetic Measurements held in Boulder, Colorado from June 28 to July 1, 1976. In that system the converted IF signal, derived from the microwave test signal, is summed with quadrature components of the IF reference signal. These quadrature components are automatically adjusted until their vectorial sum is of equal magnitude and opposite phase to the converted IF signal. The procedure involves establishing successive estimates until the balance is achieved, whereupon a measure of attenuation is found from the ratio of reference signal adjustments required to achieve balance with and without the device under test in the line. The procedure requires use of a minicomputer and is very slow and cumbersome.
It is therefore an object of the present invention to provide an improvement in the IF substitution method of microwave attenuation measurement which permits simple and accurate measurement over a range of 0 to 120 dB.
It is another object of the present invention to provide an IF substitution method for measuring microwave attenuation and/or phase shift which is devoid of the aforementioned prior art problems.